跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2611.22806 · v1

Virtual Double Oracle: Faster Convergence by Leveraging Strategies That Were Watching All Along

首次提交 2026-11-17 22:00
版本更新 2026-11-17 22:00
雷达观测 2026-11-18 19:41
arXiv ↗PDF ↗

摘要

Double Oracle (DO) and PSRO approximate Nash equilibria in two-player zero-sum games by iteratively expanding a strategy population with best responses supplied by an oracle. Online Double Oracle (ODO) interleaves multiplicative weights updates (MWU) with population discovery and obtains an O(kln⁡k/T)O(\sqrt{k\ln k/T}) rate, where TT is the number of learning rounds and kk is the larger of the two players' final population sizes. We identify the reinitialisation step as an underexplored design choice: at each expansion, MWU-based methods reset newly discovered strategies to a uniform prior. We introduce Virtual Double Oracle (VDO), which exploits a bilinear identity (a strategy's full hindsight loss history collapses to its utility against the time-averaged opponent) to initialise each new strategy with the MWU weight it would have had from the start. We prove that VDO achieves O(k/T)O(\sqrt{k/T}) exploitability over the discovered population as it grows, provided the cumulative cost of adding strategies is at most linear in kk, a condition supported by our empirical measurements. Discounted VDO (dVDO) moderates historical preferences to reduce over-concentration at reinitialisation. Experiments across matrix games, poker, and Goofspiel show that, with a suitable discount, dVDO can substantially reduce finite-horizon exploitability relative to both VDO and uniform-reset ODO, and can improve the discovered strategy populations.

作者备注:Accepted at NeurIPS 2026. 25 pages

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

目前只收录了一个版本。收录下一版后,可以对照标题、作者、分类和摘要。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v1 · 2026-11-17 22:00 · 最新观测版本

Virtual Double Oracle: Faster Convergence by Leveraging Strategies That Were Watching All Along

Double Oracle (DO) and PSRO approximate Nash equilibria in two-player zero-sum games by iteratively expanding a strategy population with best responses supplied by an oracle. Online Double Oracle (ODO) interleaves multiplicative weights updates (MWU) with population discovery and obtains an O(kln⁡k/T)O(\sqrt{k\ln k/T}) rate, where TT is the number of learning rounds and kk is the larger of the two players' final population sizes. We identify the reinitialisation step as an underexplored design choice: at each expansion, MWU-based methods reset newly discovered strategies to a uniform prior. We introduce Virtual Double Oracle (VDO), which exploits a bilinear identity (a strategy's full hindsight loss history collapses to its utility against the time-averaged opponent) to initialise each new strategy with the MWU weight it would have had from the start. We prove that VDO achieves O(k/T)O(\sqrt{k/T}) exploitability over the discovered population as it grows, provided the cumulative cost of adding strategies is at most linear in kk, a condition supported by our empirical measurements. Discounted VDO (dVDO) moderates historical preferences to reduce over-concentration at reinitialisation. Experiments across matrix games, poker, and Goofspiel show that, with a suitable discount, dVDO can substantially reduce finite-horizon exploitability relative to both VDO and uniform-reset ODO, and can improve the discovered strategy populations.

查看此版本解读 →查看此版本原文 ↗